Shoot Multiplication and Rooting Induction in Carica pubescens Lenne & K. Koch (Mountain Papaja).

1-

r--

L-r-al

L-fEdl

L-

r-.-

L-

r--

L-r

.--

L-

r,--

L-rr

fr-

L-tr

g

Proceedinq

MfE#IMf,ERNAM

@NffiRENGEffi@M

lssN

2302-108X

VOLUME

5

E

$

_ESYO

U

c{o

_N

sll

=l

r,

M

RGE

G@GRGINq$

L-f---a

L]

l--a

l_

_L.-a

I

f-r-a

l_

l-I

l_

l-a

l_

r-a

t_

l-I

l_

t{a

lj

_L{

t--L

.L

6) q fs

tr E'

o

_\J

)L

4o,

February

12

-

13,

2014

Batu, East Java, lndonesia

Faculty

of

Mathematics and Natural Science

The sih lnternational Confelence on Global

Reoutre

Gonservatbn

_I

February 12-B"l2Ol4

MESSAGE

TROM

RECTOR OF

BRAWIJAYA

UMYERSITY

Honorable Keynote Speaker

Prof. Dn ltdasrshi

Krto,

Dep. of Applied Biological Chemistry, Meiio University Nagoya

Prcf. Dr. rcr.

Nat

Helmut

Erdmann, Flesnburg University of Applied Science Pnof.

Dn

Maketrb

Mohamed, Malaysian Nature Society, Malaysia

"Er.tr'arah

Atnt

Abdullth,

School of Mathematical Sciences, Universiti Sains Malaysia Prof.

Dn

Chow Son Chen, School of Earth Science, National Central University, Taiwan

Dean Facutty of Mathematics and Natural Sciences And

All

participants

Assalamualaikurn Warahmatullahl Wabarahatahu

First of all let us pray to Allah ttrc Almighty for His blessings bestowed to all of us fiat today

lve o8n all be here to attend the The 4th annual Basic Scionce International Conference and the Sft Interndional Conference of Global Resources Conservation.

It is

indeed an honor

md

privilegc

for mc

to

welcome you

warrrly

at

ftis

Conference.

My

great appreciation

to

all

of you the distinguished participanb

of

&is important

event, and

trust

tlnt

this conferencb

will

be an valuable

input

to the earpowerment

of

Applying science

to conserue our nature.

I would like to take this opporhmlty to offer my appreciation to

Dean

Faculty

of

ldathematics and

Natural

Sciences, and committee from

Biology

Deparfrnent

Univcrsity of Brawijaya who have otganizedthis Seminar.

Dlsfrngulshed Gues&,

Itdtcs

and Gentlenun,

University

of

Brawiiaya

(UB)

as oneof the leading university in Indonesia

with

its mission

of

lVorld Class University, should take actions to participate in conserving our nature where we are tiving inside through innovation of science. We encourage all of academician here, as the

backbone

of

nation

building

for oontinuous leaming

to

save our planet

for

best interest

of

- _{human being and}

_living

_thing

_{Thereforg We should wort togedrcr, across institrtions and} aoross discipline. We should start thinking how

to

be leader and conhot the world

with

our resources, knowledgs and technoloryto achieve an equitable welfare.

Dlstiagutshed

Gurstr,,

ry

and Gentlemcn,

UB

wi&

combination betrreen International standard and local culture has been educating communities

wto

have made positive impacts

in

their

Communities-throughout Indonesia Hence

we warmly

welcomes collaborative worts

of

mutual parhership

with

many other institutions; especially universities, industry, govetnnrent and other

in*itution; botr

at national

md

international

level

Moreover

{rB

has dedicatd

itselfto

be a world class uriversity. Based

on spirit we

belief

ftat

only

with

intemational

partnenhip

able pursue multinational

February

12.13d2014

.,1

fn"

_{5m lnternatlonal Gonfelenoe}

_or

_Gbbel

_Rerource

_Gonsewdion

connectivity on business, established in the higher education institution, for future generations.

In this case, this occasion is really necessary to initiate cooperation beyond national border to

supply comprehensive knowledge to be a winner in the global competition.

DWingulshed

Gucs*, Ladies and Gentlenun,

Science, technology and education

will

determine the well-being of people and nations in the

future.

Therefore, academician and scientist as Scholar who

will

bear the future of ftris nation

have

to

worlc

har4

and

the

ability

to

create

brilliant

innovation

to

supply environmental

technolory

to solve

human problem.

UB rely

on

and

encourage increased international

parfirenhip,'as

well

as greater

staff mobility. The

parhership,

it

will be

able

to

supports

economic and social developurbnt.

In

Anothcr word, whatever challerrges that

will

arise,

it

should be challe,nges for all of us, and it should be solved with partrership.

As a conclusion, I,propose

to

stneng$ren our collaboration

to

crEate synergism at any levels,

espocially all stakeholders,

localln

regionally, as well as intenrstionally. By working togedier

all

over tlre world tbrough straring information and resources, we can make the bright future. Finally, letushope

trat

fte

frrdtful poina

aimd

from this confereirce

will

develop

tre

new

concept and networking bas€_{d on} science and technolory to save our nature.

'

'

During the organization and execrfion

of

this confer€nce, and the dne

I

am currently opening *The 4th annual Basic Science International Conference and the 5th Internafional Conference

of Global Resources Conservation".

Ladies and ge,nflemen,I hereby wish you a

fruitful

Confer,ence.

Vabfltalrto&roolhiloyah

WosatmualathunnwruhnffiullolttroofuaWuh

Thank you for your attention. Malang Fobnrary l5m, 2014 Rector,

February

12-13ttl2014

_I

*"

56lnternatlonal confercnce

on Globet Reeource

Gomenrrtion

PREtr'ACE

All

praises are due

to

Allatu God Almighty, Who made this annual event

of

successful

of

"

The *4fr

Annuel

Basic Schnce

Intemstional

Confercnce 2014

tn conjandlon

r,ffIr The Sft

Intcrnational

Conference on

Global

Resource

Consenation2Ol4, both

annual scientific

events organized by the Faculty of Mathematics and Natural Sciences, Brawiiaya University.

In

this year, the conference took a theme

of

'Applying *l"rr""

lo

"oo.u*

neture".

lh"r"

conferences are concerned about our current challenge on how

to

explore,

utilize

andapply

our

knowledge

and

science

to

consewe

water,

soil,

earth,

air,

plants, animals

and microorganisms that involved multi disciplines. As a oonjunctive conferences, these covered a

wide range oftopics on basic sciences: physicq

biolory,

ohemistry, mathematics and statistics

as well as conserryation

biolory

and applied science.

The confercnce in 2014 was the continuation of the

precding

conferences initiated in

20ll

as

the Internati,onal Confenence

on

_{Baslc Sclencc OCBS) and}

Intemstionel

Conference on

Global Rcsource Conscmatl,on inidnde in 2010. Thercforc the proceeding was also divided

into two bool6, each wittU each with a different ISSN. The proceedings were also published

in

electronic forms &at can be rccessed from BaSIC website.

I

am glad that for the first time both types

of

publication can be realized. These international

conferences

ale

held

to to

increase dissemination

of

applying science

to

conserve natural

resourc€_{s, to} present new rcsearch findings, ideas and informations and to discuss topic related

to conference theme and to develop collaboration among

multi

discipline sciences and to

find

pote'ntial

young

researchers.

This

is in

line with

university

vision as

a

World

Class

Entrepreneurial University

I

am grateful to all the members of the program committee who confributed for the success

in

farrring

the program.

I

also

trunk

atl

ttre delegates wtro contributed

to

the success

of

this

confenence by accesing our invitation and submitting paper for presentation

in

the

scienffic

prograrn.

I

am also indebted

to

PT. Fajarmas, PT. Makmur Sejati, KPRI,

CV.

Gamma, PT.

Tata Bumi Raya and CV. Enscval Medica Prima for their sup,port in sponsoring this event.

I

hope that you enjoy in this seminar. We wish for all

of

us a grand success in our scientmc

life and forthe coming conferences and even

better.

!

Thank you,

Amin Setyo Lelrsonq Ph.D.

E

tr

tr

€

l-H

l-€

Tlro 5o lnternatlonal Gonferunce on

Gbb.l

Recource

comervauon

I

February

12"8i{2014

PrrydnC.ommittee

Stcerftgcmmittce

Rect6, University of Brawiiaya

Deq

Faadty ofltde&ematics

edNdural

Sciemceq

Associato Deam

l,

2 and 3, Faculty of lvlatherretics and Natural Scicnces

Chairperson

AminSetyoLeksouo,S.Si.,M.Si.Jh.D.

(

Deputy-Ctsir

Lukman HakirL S.Si.,MSc.Dr. Sc.

Secrdary

D.

SriRatra:/rtM.Kes.

Treasurtrs

TriArdpti,MAgr.fh.D.

Mrs. Sri Punrqini

Mrs. Rusikri Adiningrum

Publication & nooscaings

Muhaimin

Rifa'i

Ph-D.

Md.Sc

Luchmm Hakino, MAgr.,Ph.D

Secrctariats

andrqifidion

j

IUrs. SriAsrinin

lvlrs. Trivira Meirany

I\,Is.

fuikArubil

IvIr.

Elsan

Scientific sections

ud

progrm

Dr.k

MocL Samito

Djt,

trtS.

Dr. Suharjono

fur.

Pori"s"noilo

Transportatioq cxcursion & social Evmt

D.

AguB Pramea

f,rrih

!6aha&a, M.Si.

Dr. Nunungllarijati

Dr.

SriWidfrti

Brian Rahadi, S.Si., lvlSc.

lvfrs.

Arnawati

'ee'

Veirue

Dr. Soff Permanq lv[.Sc, D.Sc.

Drs. Ads Socrvmdo,

I[Si.

Ifr.I(ar1,adi BkaPutra Mr. AggnsKumiawan Mr. Ifasan Muhajir Cmnmpions

Dra. Gu$iniEkowui

D.

Scrafnshlndriy€_ni

G

Februirry

t2-t3ld:mle

I

nrc

so

tntcnrrttonet Confecnce

on GloDel Recourca

Conaervdion

t-

--I

E

l

I

tE

Prof. Dr. rer. Nat Helmut Erdmann, Flesnburg Universiry of Applied

Science

"

i

Prof. Dr. Maketab Mohame4 MalaysianNatire

Society,ltalaysia

_.#

Prof. Dr. Chow Son CherU School of Earth Scicncg National Central University, Taiwan

Prof.

₎

Fatchiyalu M.Kes., Ph.D, Deparfinentof

Biolqry,

BrawijayaUniversity

g

B.Farah Aini Abdullah, School of Mathemdical Sciences, Universiti Sains

Malaysia

₎

Dr.h. Estri laras, MSt Sc.,

DepartnentofBiologr,

BrawliayaUniversity

_'t;'.'|

Dr. Sasmito Djati, Deparbnent of Biology, University of Brawiiaya,

Indonesia

)

Dr. Setijono Samino, Department of

Biolory,

University

ofBraw[iayg

Indonesia

g''i

Dr. Chomsin S. Wldodo, Department of Ptrysic, Unlverslty of Brawljaya,

lndonesia

)

-r

\7i

)

Fr

\Y.

)

-r

v.r

)

H

)

-r

=/

II

:/

H

J

-a

'yT

J

H

=T

I.{

_YI

J

-{

:I

-=1

-!?r

)

Fl

I

lt

:I

trl

vt

)

E

)

-l Vr

)

H

_V/t

)

Qrr,l

)

-i

_-, 1

)

h

")

--:/

/

7

krternational Scientific Committee and Editors

Prof

Dr. Masashi Kato, Faculty of Agriculture, Departnent of Applied Biological Chemistry,

Me[io University Nagoya

Dr. Endang Arisoesilanlngslh, Department of Biology, University of Brawljaya, lndonesia

Dr. Retno Mastuti, Department of Biology, Univerclty of Brawijaya, lndonesia

Dr. Suklr Maryanto, Department of Physlc, Unlversfi of Srauriiaya, lndonesia

Dr. Didlk R Santoso, Department of Ptrysk, Unhrenlty of Erawijaya, lndonesia

Dr. Ratno Bagus, Department of Mathematics, Unlvercity of Brawijaya, lndonesla

Dr. Wurpnsari Muharini

(,

Oepartment of Mathematks, University of Braudjaya, lndonesia

Dr. Barlah Rumhayati, Department of Chemistry, Universlty of BrawUaya, lndonesia

Dr. Elvina, Department of Chemlstry Unlverclty of Brawljaya, lndonesia

Dr. Masrurl Department of Chemistry Univercfi of BrawUaya, lndonesia

Dr. Amln Setyo Lekono, Department of Birclogy, Unfuercfi of Brawliaya, lndonesia

=,

L

TABLE OF CONTENTS

VOLUME 5

Relationship Model Anomaly Harvested Rice With a Weighted Rainfall Index

in Buru Maluku Using Bootstrap Aggregating Mars Methods To Predict The

Forecast Error Rates Harvested Area And Rice Production

……….

Kondo Lembang, Loupatty, and Talakua. M.

1

Sequential Pattern Discovery Of Deoxyribonucleic Acid (DNA) For Cancer

Patients Using Prefixspan Algorithm

………

Lailil Muflikhah,

Ilham Y Uliantoro,

and Widodo

7

Investigation on

Leaves’ Response

of

Putri Malu

(Mimosa

Pudica

L.) Respects

to Light Intensity

………...

Bagus Haryadi, Pramudhita A. Nugraha, Danu Taspyanto, and Hariyadi

Soetedjo

13

Productivity and Stability of Cassava Promising Clone ………

Sholihin

21

Shoot Multiplication and Rooting Induction in

Carica Pubescens

Lenne & K.

Koch (Mountain Papaja)

………

Enni S. Rahayu Noor A. Habibah

Mufidah H. Noviana

24

SODIUM CARBOXYMETHYL CELLULOSE (Na-CMC) FROM CORN

COB

………...…

Sirin Fairus, Dyah. S. Pertiwi, Dony Aryandi, and Wildan Priyogo

29

Automation Systems for Regeneration Process of Demin Water to Maintain

Quality Control Using Programmable Logic Controller

………...

Syahril Ardi, Sirin Fairus, Vionita Kartika Putri

34

Development of Somatic Embryos and Secondary Somatic Embryos of Cassava

……….

Surya Diantina, Darda Efendi, Ika Mariska

38

The Effect of CO2 Concentration in Modified Atmosphere Package and Storage

Length on Meat Quality of Broiler

………...……….

Herawati

43

RESOURCE POTENTIAL OF SNAKE FRUIT (Salacca

Zalacca

Var

Amboinensis) AND CANARY (Canarium

Amboinense)

IN THE LIFE OF

SERAM ISLAND SOCIETY, MOLUCCAS

………

Marcus J. Pattinama1, Aryanto Boreel, Jane K. J. Laisina, And Handy E.P.

Leimena

51

Inventorization of Edible Macrofungi From the Tropical Rainforest Ecosystem

of Meru Betiri National Park East Java

………

Arif Nur Muhammad Ansori, Laifa Fusvita,

Nastiti Trikurniadewi, Fadilatur

Rahmaniyah,

Elvin Haris Arizal, and

Ni’matuzahroh

COMPARATIVE STUDIES OF THE WATERBIRD DIVERSITY FROM

FAMILY SCOLOPACIDAE AND CHARADRIIDAE ON WONOREJO

CONSERVATION AREA IN SURABAYA

………

Alexander K. S. Putera

Nurul A. Dianti, Elvin H. Arizal, Riris Damayanti,

Robbyke Ogistyawan F., And Bambang Irawan

58

Willingness to Pay to Avoid Deforestation and Forest Degradation:




Evidence

from Berau District Indonesia

………

Akhmad Solikin

61

Identification of Vegetation Diversity for Keeping the Quality of Slope Around

Dengkeng Watershed in Klaten Central Java

………

Maridi ,

Marjono,

Alanindra Saputra,

Putri Agustina

65

DIVERSITY INDEX OF HOLOTHUROIDEA (ECHINODERMATA) AT

CASED TRADITIONAL GOLD MINING AREA, LAMPON BANYUWANGI

DISTRIC

………

Susintowati, Nurchayati. N

70

Diversity of Ophiuroids in Intertidal Zone of Porok and Kukup Beach Gunung

Kidul Yogyakarta

………...

Fibra Nuraeni, Donan Satria Yudha

75

Increasing of Soil Enzyme Activities by Addition of Yeast Extracted from

Semi-Solid Waste of Alcohol Industry

………

V. Irene Meitiniarti, Rully Adi Nugroho, And E. Maskito

79

Population Dynamic of Parasitic Plant

Viscum

Articulatum

Burm.F. in

Purwodadi Botanic Garden

………

Solikin

82

The Influence of Explant Sterilization Method and 2,4-Dichlorophenoxy Acetic

Acid (2,4-D) on The Callus Formation of Various Explants of Red Betel




(Piper

Crocatum)

………

Junairiah, Ni’matuzahroh,

Hery Suwito

87

“BORNEO PRIMA”, MANDARIN CULTIVAR FOR LOW A

ND WET AREA

FROM KALIMANTAN INDONESIA

……….

Titiek Purbiati, Arry Supriyanto

91

PUERARIN INHIBIT VASCULAR CELLULAR ADHESION MOLECULE-

1(VCAM-1) AND TUMOR NECROSIS FACTOR-

Α (TNF

-

Α) EXPRESSION

ON HUMAN UMBILICAL VEIN ENDOTHELIAL CELLS (HUVECS)

AFTER INDUCED BY LEPTIN

IN VITRO

………

M. Sasmito Djati, Retty Ratnawati,Satuman, Erly N. Aisyah, Ratna Semestiari,

Rasyad Indra

Sea Holy (Acanthus

Ilicifolius)

Viability Test on Heavy Metal: Lead (Pb) and

Cadmium (Cd)

………...

Rony Irawanto

February 12-13

2014

The 5 International Conference on Global Resource Conservation

Batu, East Java, Indonesia

| 1

Abstract— Seasonal climate variations is one of the main

causes of the diversity of crop production in Indonesia. Long drought and drought causing crop failures and food shortages that could affect agricultural production and food security. The indicator is a decline in acreage, harvested area and production declined sharply when climate irregularities. The magnitude of the impact caused by climatic irregularities cause we need a model that connects the harvested area with indicators of climate anomalies that can do the proper planning and anticipation measures early in order to avoid the risk of crop failure. Buru as the largest rice -producing areas in the provinces of Maluku course is expected to avoid the risk of crop failure in order not to disrupt the supply of rice. Data Collection and forecast rice production annually conducted by the Central Statistics Agency (BPS). BPS forecast model but has not entered a climatic factor, while the climate affect rice production. This research used the bootstrap aggregating MARS method to model anomaly rice harvested area with a weighted rainfall index to predict the error rate forecast harvested area and rice production. From the analysis using the best models of replication bagging MARS 150 times in the first period (January-April) and 200 times in second period (May-August) and third period (September-December) obtained an error rate forecast harvested area and rice production respectively by 5.72 % and 6.81 % .

Keywords— Anomaly Area harvested, weighted rainfall index, MARS, Boostrap Aggregating, rice production

I. INTRODUCTION

ice is the main food for the people of Indonesia, which provides seasonal income and employment for rural communities. Rice production has increased since 1970, but the harvest is particularly vulnerable to climate variability of extreme events: El-Nino and La-Nina. In the event of the El - El-Nino rice production has decreased quite dramatically, as in 1991, 1994, and 1997. Similarly, in the La-Nina (1995) also decreased rice production [8].When the national rice supply is insufficient and a decline in production, the import policy is often done. The problem is the need to

forecast future production decline (extreme weather events), so the anticipation can be done. That requires models that accurately forecast rice production in order to support national food security. Data Collection and forecast national rice production every year conducted by the Central Statistics Agency and the Ministry of Agriculture. Forecasts made by the province were calculated based on time series data and the provinces are not based on the sum of the forecast district/city level. Production per province is obtained by multiplying the harvested area clean with a yield per hectare per unit of harvested area clean for every sub round (4 monthly: sub round 1 (January-April), sub round 2 (May-August), and sub round 3 (September-December) [5]. production and harvested area in a year (January to December) is obtained from the sum of production and harvested area for three sub round. Yield to each hectare is the yield each hectare in the form of tile results each unit of harvested area.

II. REVIEW OF LITERATURE

2.1 . MARS

MARS is an implementation technique popularized by [4] for solving regression problems with the aim of predicting the response variable values of a number of predictor variables. MARS is an approach to the development of Recursive Partitioning Regression (RPR) which still has the drawback that the resulting model is not continuous at the knots. MARS model is used to overcome weaknesses in the model generate the RPR is continuous at knots. In spline modeling, the first step is to determine the points of data or a change in the pattern of behavior is called the point knots. The selection of knots in MARS using forward and backward algorithms. The selection of the model using a forward step taken to get the maximum number of base functions with base selection criteria function is to minimize Average Sum of Square Residual. To parsimony concept (simple model) performed a backward step is selecting base functions of forward

Relationship Model Anomaly Harvested Rice

With a Weighted Rainfall Index in Buru Maluku

Using Bootstrap Aggregating Mars Methods to

Predict the Forecast Error Rates Harvested Area

and Rice Production

R

Kondo Lembang. F1), Loupatty. G2), and Talakua. M.W3)

1), 3)

lecture of Mathematic Department, Faculty of Mathematics and Natural Sciences,Pattimura University

2

The 5th International Conference on Global Resource Conservation

February 12-13

rd

2014

2 |

Batu, East Java, Indonesia

stage by minimizing the value of the Generalized Cross - Validation or GCV [4].

The minimum GCV as a criterion for determining knots are as follows:

( )

(

( )

)

( )

2 1 2

1

ˆ

1













−

−













=

∑

=

N

M

C

x

f

y

N

M

GCV

N i i M i where

M

: The number of base functions

( )

M

C

: The number of parameters in the model =trace (B(BTB)-1BT)+1

B : matrix of base function

N

: The number of data

i

y

: Value of the response variable

( )

i M

x

fˆ

: Estimated value of the response variable on

M

base functions.

From forward and backward, MARS models obtained as follows:

( )

∑

∏

[

(

( )

)

]

= = +

−

+

=

M m K k km m k v km m m

t

x

s

a

a

x

f

1 1 , 0

ˆ

where 0

a

: main of base function

m

a

: coefficients of m-base functions

M

: maximum of base function

m

K

: degree of interaction

km

s

: ± 1

(km)

v

x

, : independent variables

km

t

: Point knots of independent variables

(km)

v

x

_,

Algorithms for MARS models are as follows:

1. Starting with a simple model involving only constant base functions.

2. Finding space of base functions, for each variable and for all knots are possible, and add it to minimize the prediction error.

3. Repeat steps 2 to obtain a model that has maximum complexity.

Finally, in the last stage, the trimming procedure is applied in which the base functions are not significantly removed to obtain the minimum GCV.

2.2. Bagging Mars

Bagging method was first used by Breiman (1994). Bagging is used as a tool to form a more stable classifier. Bagging predictors is a method to generate multiple versions of a predictor and use it to aggregate predictors. Multiple versions of the bootstrap replication is formed by a set of data.

Defined a set data £ consists of

(

y

n

,

x

n

)

,

n

=

1

,

K

,

N

where

y

a numerical response

or a class label. If x the input is then

y

predicted by

ϕ

(x,£), where

ϕ

(x,£) is a predictor. To gain a better predictor performed bootstrap replication {£k} is

then called {

ϕ

(x,£k) }. Performed totally B-times of

bootstrap replication so that {£(B)} from £ where {£(B)} resampling with replacement and established predictors of {

ϕ

(x,£(B)}.

Bagging MARS algorithm is as follows.

1. Taking bootstrap n samples of set data £ with n- repetitions to each aggregate variables in each observation.

2. MARS modeling sets data £B bootstrap sample

results.

3. Test the model generated in step 2.

4. Repeating steps 1-3 as much as B-times (bootstrap replication).

5. Obtain the best model.

6. Forming bagging MARS models of the average of each parameter at each sampling to B-times.

To obtain better results then the bootstrap replication is done as much as possible.

2.3. Weighted rainfall index and anomalies Harvested Rice

Weighted rainfall index (weighted rainfall index: WRI) developed in Australia by Stephen, et al (1994). This index is compiled based on monthly rainfall data is weighted. WRI which can be used in the modeling is that the WRI weighted system has been modified by Sutikno (2008). The modification is written as follows.

dard s t D t D t

L

LT

R

WRI

tan * , ,

=

where,

∑

=

=

m j j j D t

R

A

A

R

1 * , ,

∑

=

=

m j j

A

A

1 Description: * ,D t

R

: Area weighted rainfall Regional weather forecast region (DPM) / DPM revision in the region/district/city

m : a large area of DPM

j

A

: Total area of j- DPM

t

LT

: Plant area at t-month

dard s

L

tan : standard area for rice crops in the regional

j : DPM regional (1,2,3, ..., m)

t : Months (1, 2, ..., 12)

D : Regional

Model anomalies rice harvested area to each period (AnLPp)

with a weighted rainfall index (WRI) is as follows. AnLPpi = WRI1i + WRI2i + WRI3i + WRI4i

From the equation above three equations obtained for each of the following.

AnLP1i = WRI1i + WRI2i + WRI3i + WRI4i

AnLP2i = WRI5i + WRI6i + WRI7i + WRI8i

AnLP3i = WRI9i + WRI10i + WRI11i + WRI12i

February 12-13

2014

The 5 International Conference on Global Resource Conservation

Batu, East Java, Indonesia

| 3

i

= 1, 2, 3,..., n (n is the number of observations)

p = 1, 2, 3 (Period)

AnLP1 = harvested area Anomaly in first period

(January to April)

AnLP2 = harvested area Anomaly in second period

(May to August)

AnLP3 = harvested area Anomaly in third period

(September to December)

WRI1, ..., WRI4 indicates a weighted rainfall index first

until the fourth month in a first period (WRI1 = in

January, WRI2 = in February, WRI3 = in March, WRI4

= in April).

WRI5, ..., WRI8 indicates a weighted rainfall index first

to fourth month in a second period (WRI5 = in May,

WRI6 = in June, WRI7 = in July, WRI8 = in August).

WRI9, ..., WRI12 indicates a weighted rainfall index first

to fourth month in a third period (WRI9 = in September,

WRI10 = in October, WRI11 = in November, WRI12 = in

December).

III. METHOD OF ANALYSIS

Methods of data analysis performed in this study can be explained as follows.

1 . Identification of data include the identification and the relationship between WRI and AnLPp that can

be shown on the scatter plot.

2 . To model based anomaly harvested area weighted rainfall index for the data in- sample using the bagging MARS method with the following stages.

In MARS models for the first sets data

- Determine the maximum base functions

- Determine the maximum number of interactions

- Determine the minimum number of observations between knots

- Determine the number of degrees of freedom

Getting the best MARS models for the initial set of data based on the value of the smallest MSE and GCV.

Getting the significant variables of the best MARS models for the initial set of data. Perform bagging of the pair response variable and the predictor variables were significant from the best MARS models for data sets beginning with 50, 100, 200, and 250 bootstrap replication.

Perform MARS modeling on each sample-B bootstrap replication with the maximum number of base functions, the maximum amount of interaction and the minimum number of observations between knots is equal to the maximum number of base functions, the maximum amount of interaction and the minimum number of observations between knots at best MARS models for data sets beginning.

Getting MSE at each sampling B bootstrap replication.

Getting MSE bagging of the average MSE at each sampling to B

Bagging MARS model obtained is the best MARS models for the initial data sets. This is

because the value of changing each knots for each replication so that the estimated parameters can not be averaged.

3. Counting rice production forecast for the year to out -sample of data as follows.

Calculating forecast rice harvested area forecast results by adding the anomalous area harvested to each period results of MARS best modeling with an average area harvested during a certain period (2002 to 2008).

Suspect productivity to each period using the average productivity of the last five years (2007 to 2011).

Calculating forecast rice production to each period.

Pp = Prop x LPp dengan p = 1,2,3

Pp : forecast of production in the

p-period

Prop : forecast of productivity in the

p-period

LPp : forecast of harvested area in the

p-period

Getting forecast of rice production for the year which is the sum of the third period of the forecast.

Comparing rice production forecast results have been obtained with the MARS method with the actual value of rice production from BPS issued last three years (2009 to 2011).

IV. DISCUSSION

4.1. Identification Data

To model the relationship between the anomalous area harvested and predictor variables weighted rainfall index first identified patterns of relationship between the two. Identification of relationship pattern is very necessary to know the exact model in modeling the relationship between the two variables.

Figure 1. Scatterplot AnLP and WRI in Buru Figure 1 shows that the pattern of relationships between AnLP and WRI appears no linierity clear pattern not even have a specific pattern. Likewise, the direction of the relationship is positive or negative. This suggests that AnLP in first period (January to April), second period (May to August), and third period (September-December) are not affected by these two variables are linearly so necessary to find the possibility of non-linear statistical models.

The 5th International Conference on Global Resource Conservation

February 12-13

rd

2014

4 |

Batu, East Java, Indonesia

4.2. Relationship between the model anomaly rice

harvested to each Period (AnLPp) and weighted rainfall index (WRI) using Bagging MARS method.

In this study bagging MARS method is applied in modeling the anomalous area harvested per period as the response variable and weighted rainfall index as a predictor variable. The data used for modeling can be quite small, namely 10 years since year 2002 to year 2011, so it needs to be done in preparing the model resampling methods. Resampling method used is that a bagging sampling with replacement for a data set consisting of the response variable and the predictor variables (significant base functions).

4.2.1. Relationship between the model anomaly rice

harvested area (AnLP) and the weighted rainfall index (WRI) in first period using bagging MARS method.

A. MARS Model

Initial step of MARS modeling done by trial and error on the maximum base function (BF), maximum interaction (MI), minimum number of observations between knots or minimum observation (MO), and the number of degrees of freedom (DB) until an optimal model is obtained with the MSE and GCV minimum. [4] suggest a maximum number of basis functions of two to fourth times the number of predictor variables. Maximum interaction one, two, or three with a consideration if more than three will result in a very complex model. As well as the minimum distance between knots or knots as the minimum observation between 0, 10, 20, 50, and 100.

Table 1. Trial and Error Model MARS a first period in Buru

Table 1 shows the value of R2, MSE, and GCV in combination BF, MI, MO, and DB. based on the criteria of goodness of the model, selected models with minimum MSE and GCV. From the above results it can be concluded the model with a combination of BF = 8, MI = 1, MO = 0, and DB = 1 is the best model. The best MARS model is as shown below.

( )

1031

,

325

985

,

757

1

9439

,

66

4

ˆ

_x

_BF

_BF

f

=

+

−

where

From the best MARS models can be interpreted that each increase of one unit of the base functions 1 (BF1)

can increase rice yields broad anomalies in the period 1 at 985.757 if weighted rainfall index in January (WRI1)

more than 0.088 mm, with a base of other functions that go assumed to be constant in the model. Meanwhile, for each increase of one unit of the base function 4 (BF4)

can reduce anomalies rice harvested area of 9439.66 a

first period if the weighted rainfall index in February (WRI2) of less than 0.251 mm with base other functions

are included in the model held constant. The next best model obtained from two predictor variables were entered into the model, which is a weighted index of rainfall in February (WRI2) and weighted rainfall index

in January (WRI1) based on the relative variable

importance table. Percentage of contribution weighted rainfall index in February (WRI2) and weighted rainfall

index in January (WRI1) are shown in Table 2 below.

Table 2. Percentage of contributions of each variable in the first period

Variable Contribution

WRI2 100 %

WRI1 57,711 %

B. MSE calculations on models of bagging MARS

MARS modeling of the data sets obtained MSE Value in the first period is 544,216.122. To minimize the error variance performed on the data resampling. Table 3 below shows the results for the first period bagging MARS in Buru.

Table 3. Results of bagging MARS in first period

Bootstrap replication

average value of MSE

Decrease in the value of MSE

25 times 19511,6 524704, 522 50 times 3566,25 540649,872 100 times 3993,14 540222, 982 150 times 1538,90 542677, 222

200 times 2413,91 541802,212 Table 3 gives the smallest MSE value of the information obtained during the boostrap replicate as much as 150 times. thus it can be concluded that the best results obtained in the replication boostrap bagging as many as 150 times. Bagging models can lower the MSE value of the data model that is equal to the initial set of 544216.122 be 1538.90 or in other words bagging can reduce the value of MSE of 542677.222 of the initial data sets.

4.2.2. Relationship between the model anomaly rice harvested area (AnLP) and the weighted rainfall index (WRI) in second period using bagging MARS method.

A. MARS Model

Trial and error to BF, MI, MO, and DB MARS modeling in second period are shown in Table 4. Table 4. Trial and Error Model MARS a second period in Buru

BF MI MO DB R2

MSE GCV 8 1 0 1 0.261 637497.059 1147494.92

8 1 0 2 0.088 637497.059 1416660.57

8 1 0 3 0.088 637497.059 1416660.57

12 1 0 1 0.261 637497.059 1147494.92

12 1 0 2 0.088 637497.059 1416660.57

12 1 0 3 0.088 637497.059 1416660.57

16 1 0 1 0.261 637497.059 1147494.92

16 1 0 2 0.088 637497.059 1416660.57

16 1 0 3 0.088 637497.059 1416660.57

From Table 4 it is seen that the best MARS model is a combination of BF = 8, MI = 1, MO = 0, and DB = 1. It can be seen from the MSE and the smallest GCV among others, are respectively 637497.059 and 1147494.92 so the best MARS model is as shown below.

( )

566

,

237

1654

,

87

1

ˆ

_x

_BF

f

=

−

+

where

BF MI MO DB R2 _{MSE GCV}

8 1 0 1 0.507 544216.122 1523805.55

8 2 0 1 0.204 1240372.80 3088471.75 8 3 0 1 0.204 1240372.80 3088471.75 12 1 0 1 0.507 544216.122 1523805.55 12 2 0 1 0.204 1240372.80 3088471.75 12 3 0 1 0.204 1240372.80 3088471.75 16 1 0 1 0.507 544216.122 1523805.55 16 2 0 1 0.204 1240372.80 3088471.75 16 3 0 1 0.204 1240372.80 3088471.75

February 12-13

2014

The 5 International Conference on Global Resource Conservation

Batu, East Java, Indonesia

| 5

This model can be interpreted that each increase of one unit of the base functions 1 (BF1) can increase rice

yields broad anomaly in second period for 1654.87 if weighted rainfall index in August (WRI8) more than

0.22 x 10-7 mm, on the base of other functions in the model are held constant. Furthermore, Table 5 looks only variable weighted rainfall index in August (WRI8)

are included in the model. So important variable scores for weighted rainfall index in August (WRI8) worth

100%, which means the variable weighted rainfall index in August has a dominant influence on anomalous rice harvested area in second period (May-August).

Table 5. Percentage of contributions of each variable in the second period

Variable Contribution

WRI8 100 %

B. MSE calculations on models of bagging MARS

The best MARS models between anomalous rice harvested area weighted rainfall index in second period provide information that model has a MSE is 637,497.059, with a significant predictor variables are weighted rainfall index in August (WRI8). Table 6

below shows the average MSE in second period. Table 6. Results of bagging MARS in second period

Bootstrap replication

average value of MSE

Decrease in the value of MSE 25 kali 13521,9 623975,159 50 kali 15068,9 622428,159 100 kali 7378,52 630118,539 150 kali 4284,14 633212,919 200 kali 3017,05 634480,009

Table 6 provides information that bootstrap replication earned 200 times average value of the smallest MSE is 3017.05, So based on the above results it can be concluded that the results obtained by bagging the best with an average value of the smallest MSE is a bootstrap replicate as much as 200 times. With replication as much as 200 times, bagging can reduce MSE of the initial set of data models is 637497.059 be 3017.05 or in other words bagging can reduce the value of MSE of 634480.009 from the initial set of data models.

4.2.3. Relationship between the model anomaly rice

harvested area (AnLP) and the weighted rainfall index (WRI) in third period using bagging MARS method.

A. MARS Model

Table 7 gives information based on that value the smallest MSE and GCV is combination of BF=8, MI=1, MO=0, and DB=1. So that model with the combination is the best model. This model is shown below.

( )

2355

,

804

24391

,

732

2

ˆ

_x

_BF

f

=

−

Where

(

_{) (}

₎

   < − ≥ = − = 141 , 0 , 141 , 0 141 , 0 , 0 141 , 0 , 0 max 9 9 9 9 2 WRI if WRI WRI if WRI BF

Table 7. Trial and Error Model MARS a third period in Buru

BF MI MO DB R2

MSE GCV

8 1 0 1 0.761 453362.558 816052.711

8 2 0 1 0.749 453362.558 858437.928

8 3 0 1 0.749 453362.558 858437.928

12 1 0 1 0.761 453362.558 816052.711

12 2 0 1 0.749 453362.558 858437.928

12 3 0 1 0.749 453362.558 858437.928

16 1 0 1 0.761 453362.558 816052.711

16 2 0 1 0.749 453362.558 858437.928

16 3 0 1 0.749 453362.558 858437.928

From the best model can be interpreted that each increase of one unit of the base function 2 (BF2) can

reduce anomalies rice harvested area in the period 3 of 24391.732 if weighted rainfall index in September (WRI9) of less than 0.141 mm, with bases other

functions assumed to be constant in the model. Later in the third period, based on table 8 weighted rainfall index in September (WRI9) contributes 100%, which means

the variable weighted rainfall index in September to have a dominant influence on anomalous rice harvested area in the third period.

Table 8. Percentage of contributions of each variable in the third period

Variable Contribution

WRI9 100 %

B. MSE calculations on models of bagging MARS

The best MARS models between anomalous rice harvested area weighted rainfall index for the third period provides information that the model has a MSE is 453362.558 significant predictor variables are weighted rainfall index in September (WRI9). Table 9 below shows the average MSE results of bagging MARS in third period.

Table 9. Results of bagging MARS in third period

Bootstrap replication

average value of MSE

Decrease in the value of

MSE

25 kali 11418,2 441944,358 50 kali 9452,84 443909,718 100 kali 3224,57 450137,988 150 kali 1213,38 452149,178 200 kali 961,970 452400,588

Table 9 provides information that the bootstrap replication earned 200 times the average of the smallest MSE value is 961.970, so it can be concluded that the results obtained by bagging the best with an average value of the smallest MSE is a bootstrap replicate as much as 200 times . With replication as much as 200 times, bagging can reduce MSE of the initial set of data models for 453362.558 be 961.970 or in other words bagging can reduce the value of MSE is 452,400.588 from the initial set of data models.

4.3. Rice Production forecast

To evaluate the model and see the level of reliability that is formed, it can be seen from the average forecast error rates for harvested area and production of rice in the year 2009 to the year 2011. Rice production Forecast to each period is the multiplication of the value of the harvested area forecast productivity. Used to estimate the productivity of the average productivity value over the last five years . While the forecast harvest

The 5th International Conference on Global Resource Conservation

February 12-13

rd

2014

6 |

Batu, East Java, Indonesia

area was obtained from the sum of the forecast anomalies rice harvested area with average area of rice crop. Of the best MARS models which have been obtained from the previous analysis can be calculated models forecast values for harvested area and production of rice to each period as shown in Table 10. Table 10. forecast Value of harvested area and rice

production to each period (using models MARS)

∆ = Actually – forecast

Table 10 provides information based on the results of the forecast error for the harvested area and rice production to each period look distinctly average prediction error rate is at least third periods respectively 14.61% and 14.36%, followed by a period of 2 each by 20.01% and 18.97%, as well as period 1 respectively 26.53% and 26.52%. BPS and the Ministry of Agriculture every year to data collection and forecast rice production in Indonesia is divided into three periods, namely from January to April, May to August, and September to December. Harvested area of each period obtained from the amount of harvested area in the first month until the fourth month in a period. Production and harvested area in one year (January to December) is obtained from the sum of production and harvested area for three periods. So the forecast results for the year and harvested area of rice production MARS models as shown in Table 11 below.

Table 11. Forecast value of harvested area and rice Production Harvested to each year

Based on the results of table 11 forecasts to each year for harvested area and rice production of MARS models can be calculated the average error rate forecast harvested area of rice by 5.72 % while the average error rate for rice production forecast is 6.81 %. Average error rate of rice production forecast issued BPS to each province ranged from 5% to 10 %.

V.CONCLUSIONS

Results of rice harvested area forecast MARS model in Buru has an error rate forecast is 5.72 %. As for rice production forecast results have an error rate of 6.81 %. In accordance with an error rate forecast of rice production issued BPS ranging between 5 %-10 %, it can be said rice production forecast errors in Buru is in accordance with the rate specified BPS.

REFERENCES

[1] Breiman. L, Bagging Predictor. Technical report 421. Statistics department of the University of California, 1994.

[2] Efron. B. and Tibshirani. R.J, An Introduction to the Bootstrap, Chapman & Hall, Inc.,New York, 1993.

[3] Eubank. R. L, Spline Smoothing and Nonparametric Regression, Mercel Dekker, New York, 1988.

[4] Friedman. J. H, Multivariate Adaptive Regression Splines , Rev Tech Report 102, Department of Statistics Stanford University Stanford , California, 1990.

[5] Kartini. A. Y, Bagging MARS for Developing forecast Models of Anomalies Area Rice Harvested of Weighted Rainfall Index in Ngawi [ thesis ], Graduate Program, Sepuluh Nopember Institute of Technology, Surabaya, 2010.

[6] Sutikno, Statistical Downscaling GCM Outcomes and Utilization for Rice Production Forecasting (Disertation), Bogor, Graduate Program, Bogor Agricultural University, 2008.

[7] StatSoft ., Multivariate Adaptive Regression Splines (MARSplines) , Copyright StatSoft , Inc, 2000.

February 12-13

2014

The 5 International Conference on Global Resource Conservation

Batu, East Java, Indonesia

| 7

Abstract— The high rate of mortality is caused of cancer

disease due to it is effect of uncontrolled growth of cells and mutations. It effects many research to be conducted for early detection of the desease. One of them is a study and analysis of the molecular level is done by combining molecular biology, mathematics and informatics which is known as bioinformatics. A method in the field of informatics that can be used to find the pattern of Deoxyribonucleic Acid (DNA) sequence databases as a constituent amino acids of the p53 gene is called Sequential Patterns Mining Discovery. This method is a process of extracting data to generate knowledge about the series of events that has the appearance of a certain frequency. This research is proposed PrefixSpan algorithm to discovery the patterns due to it has a high performance in computation time. The test is implemented using several threshold, such as minimum support (minimum frequent of sequence in the whole transaction) and sequence length in each transaction database. It takes from DNA sequence databases of cancer patients in each the same exon. The test result is obtained patterns or motifs of DNA sequences for the patients. The performance measure shows that the average support is high and stable in the range of 0.8. This number means that the frequent of thus pattern is high. Another performance measure is lift ratio which has average value more than 1 . This value shows that the generated patterns have high relationship and dependency. Keywords— cancer, bioinformatics, DNA, sequential pattern, PrefixSpan

I. INTRODUCTION

ANCER is classifed as mallignant and deadly disease. Generally, it is caused of gene mutation, i.e. p53 and it effects to change of p53 protein sequence [1]. This protein consists of a combination of 20 amino acids which are synthesized by ribosomes are formed based on the genetic code of the Deoxyribonucleic Acid (DNA). If the DNA is mutated, then the protein composition will be incorrect. Continuously, it is effect to variety of diseases and disorders such as cancer. Therefore, early detection can be conducted by analizing of protein sequence in the blood test. The most frequently altered gene of P53 or TP53 mutations is found in human cancers. There are 30,000 somatic mutations of various cancer types in TP53 database which is collected over 20 years. Recently, the methodology of sequencing cancer genome impacts on the healing and data management

[2]. According Soussi, p53 mutation analysis of the pattern has become essential to investigate the cause of cancer. His test result shows that infrequent of gene mutation is associated with the normal activity of the p53 protein [2].

The high rate of mortality which is caused by cancer, it makes early detection of this disease. A field that is expected to provide his role is bioinformatics. This field is a knowledge discipline that combines the study of molecular biology, mathematics and information technology (IT). It is defined as the application of computational tools and analysis to capture and interpret molecular biology data . Molecular biology itself is also an interdisciplinary field in molecular level of life sciences [3]. Bioinformatics has a very important role, including the data management for molecular biology, especially DNA sequences and has huge volume of genetic information. One of IT field that is used to find the DNA patterns of cancer patiens is Sequential

Pattern Discovery. Sequential pattern discovery is a part

of data mining task that generates knowledge about the series of events that have a frequency of occurrence that exceeds a specified threshold value [4]. The pattern is expected to use as biomaker of cancer disease. This research is proposed PrefixSpan algorithm to discovery its pattern of DNA sequence database. This algorithm is a method of Sequential Pattern Discovery which has high performance of computational time [5].

II. SEQUENCING GENOME

DNA sequencing is a process of determining three million nucleotide bases order which consist of adenine, guanine, cytosine and thymine (A, T, G ,C) in a DNA molecule. However, sequencing genome is the determination of the nucleotide sequence of DNA bases in the genome or in the body of an organism . Sequencing results are expressed in the form of a sequence of letters nucleotide bases in particular DNA, for example AGTCCGCAGGCTCGGT. Sequencing genome is always compared to coding process, whereas the sequencing process is not only defining a code, but also analogous genome sequence of letters from a mysterious language. It has an important and specific meaning.

Sequential Pattern Discovery of

Deoxyribonucleic Acid (DNA) for Cancer

Patients Using PrefixSpan Algorithm

Lailil Muflikhah 1*), Ilham Yuliantoro 1), and Widodo 2)

1)

Program of Information Technology and Computer Sciences

2)

Faculty of Mathematics and Natural Sciences

1, 2)

Brawijaya University, Malang, Indonesia

*)

lailil@ub.ac.id, laililmf@gmail.com

The 5th International Conference on G

Furthermore, the sequence alignment is of setting two or more sequences so that the similarities can be detected. Sequence align interpolated (usually with the sign "-") s columns contain characters that are identic between the sequences. Here is an exa alignment of short sequences of DNA different DNA, "ccatcaac" and "caatgggc indicates a match or matches betwee sequences) are shown in Fig. 1.

Fig. 1. Matching two sequences The mismatch in alignment is assoc mutation, but the gap (the sign "-") is asso insertion or deletion. In addition, sequence also used to find similarity in sequence data III. DATA MINING Data mining is a method of multiplying obtain the hidden information. Defining a w data mining [7]:

• It is an untrivial decomposition of potential information implicitly.

• It is mining and analysis by an semi-automatic utility from huge patterns that have meaning.

• Data mining is also part of the discovery in databases (KDD) Briefly, there are two methods of da implement the role, include [6]:

• Prediction Method

It is using some variables to pred values or future values of other var

• Method Description

It seeks human-interpretable patte explain certain data.

3.1. Sequential Pattern Discovery

Sequential pattern mining is a data mining generates knowledge about the sequence o have an occurence frequency that exceeds value [4]. Sequential pattern is a derived pa association rules, because both indicate the between events. The difference is that the pattern of events focused on finding patter that appear after another event, but associat pattern of events that occurently appear events.

3.2. PrefixSpan Algorithm

PrefixSpan (Prefix-Projected Sequent Growth) is a development approach th algorithm to search for sequential patter PrefixSpan will seek frequent sequences of and then develop the sequences by adding by one. As a result, the additional sequenc

n Global Resource Conservation

February 12

is the process the sequences lignment line is such that the tical or similar xample of the NA from two gcaac" (the "|" een the two

ces

sociated to the ssociated to the ce alignment is databases.

ng the data to a wide range of on of dataset into n automatic or ge data to find the knowledge data mining to

predict unknown variables. attern that can

ing process that of events that ds the threshold patten from the the relationship the sequential tterns of evennt iation rules is a ppear with other

uential Pattern that uses an tern sequences. of the elements elements one uence is still the

previous sequence. This way is not generate and test for candidates. The PrefixSpan algorithm is shown in Fig. 2

Fig. 2. Pseudocode of PrefixSpan Alg 3.3. Evaluation

In order to know the performance of syat some measurements such as support, conf ratio and accuration rate.

• Support

It is probability of frequent itemset transaction. It is ratio of transaction of an itemset as shown in Eq. 1[4].

∩

(1) • Confidence

It is measurement which shows rela two items conditionally. Confidenc 2.[4].

! "!, → %

Where:

• & ∪ = the number of transactions

• & = the number of transaction

• Lift ratio

Lift ratio is a measure to know constructed rule of sequential algorithm. The value of lift ratio is unlimited. It can be defiined as below

- If lift ratio value is less means that rule antacedent negative of rule consequent. - If lift ratio value is equal to is frequent but it is independ - The others, if lift ratio value then it means that the rule

Input: sequence database, minimum Output: A complete set of sequent Method: Call prefixspan({},(),S). Subroutine: Prefixspan(α,l, S|α)

Parameter: α: suatu seqential pat length α,

S|α: projected database, if α ≠()

then sequence database is S

Method:

1. Scan S|αonce, find set frequ

so that:

a. b can be combined to las

α to construct sequentia b. <b> can be added to α fo

squential pattern 2. For each item b is appear, a

constructing a sequential pa output α’;

3. For each α’, construct α‘ pr database S|α’, and call Prefi

l+1, S|α’).

uary 12-13

rd

2014

not necessary to he pseudocode of 2 [5].

lgorithm

yatem, there are confidence, lift

et in whole ion which consists

].

( ) * +

elation between nce is stated is Eq

% )∪+ % ) (2) of itemset in all

of antacedent in

w the strength of l pattern mining is between 0 and elow: [8]

s than 1, then it nt will be effect to

nt.

to 1, then the rule ndent.

lue is more than 1, le antacedent will

m support tial pattern .

ttern, l:

(); if it is not

uent itemset, st element from al pattern; or or constructing add to α for attern α’, and

rojected ixSpan (α‘,

February 12-13

2014

The 5 International Conference on Global Resource Conservation

Batu, East Java, Indonesia

|

97

Abstract— The development become in increased human

activity in various sectors, such as: industrial, residential, agricultural, and other sectors. Human activity in the development that not consider the environmental aspects, impact on improving the quality and quantity of waste that will contaminate various environmental parts. Heavy metal pollution is a serious attention, if absorbed and accumulate in the human body can be disorder to health and cause death. Heavy metals are toxic pollutants in all environmental parts, such as Pb (lead) and Cd (cadmium). Plant species of Acanthus ilicifolius an annual shrub, woody, single leafy, spiny, flowering compound, fruit ovoid, contain four seed each fruit. Distribution in the tropics, especially Indonesia, often found along the edge of the river with the sea, and in the mangroves near the coast. Can grow to 500 m above sea level. The study aims to determine the viaability of Acanthus ilicifolius on heavy metals (Pb and Cd). Research conducted at the laboratory scale in greenhouse of Environmental Engineering - ITS. Methods with range finding test use three replications, observation gradually over two weeks to determine the letal concentrations of heavy metal. Result showed that

Acanthus ilicifolius cant live in concentration for lead ( Pb

) and cadmium ( Cd ) is 8000 mg/L and 300 mg/L.

Keywords— Seaholy (Acanthus ilicifolius), Lead (Pb),

Cadmium (Cd).

I. INTRODUCTION

UMAN population growth and development resulted in increased human activity in various sectors development, such as industrial, residential, agricultural, and other sectors. The increasing human activities in various sectors, often increasing the amount of waste pollutants. Wastewater pollutants that enter into the environment can disrupt and harm the environment. The impact of environmental pollution, both directly or indirectly affect the human body health and cause systemic disorder. Pollutants enter the human body through contact with the skin, respiratory and digestive organs. Therefore, if the pollutant is not managed properly will cause environmental pollution, endangering the health and life, both human and other organism.

Heavy metal pollution is a serious attention, because can absorbed and accumulate in the human body can be

detrimental to health [1] and in some cases cause death.

Heavy metals can accumulate in the environment and moves from one medium to another medium [2]. So heavy metal can contaminants in all environmental media (multi-media pollutant). Heavy metals are not needed by organism because it is toxic such as Pb (lead), and Cd (cadmium). Pb and Cd has a widespread distribution and a major cause of environmental pollution and health problems [3].

The application of science and technology to assess and prepare solutions to environmental problems using plants known as Phytotechnology. Phytotechnology used to expand understanding of the importance of plants and their role in human life and environmental systems. The phytotechnology concept is focused on plants as environmental technology that able to solve environmental problems. In a review of the technology and process, phytotechnology clarify show nature-based approach to environmental problem solving. This is the balance technology between man-made process and natural-plants process, as representation of how the two processes to overcome various environmental problems [4].

Phytotechnology study based on transformation of substances in the ecotoxicological effects. Ecotoxicology have common with the study of environmental toxicology in products such as the negative effects of substances on living organisms and treatment outcome for substance restrictions. However, the difference in the method of exposure substances, size substances, carrier substances and objects organism. So that needs to know that the negative effects of a substance as health preservation and sustainability of life organism [5].

Ecosystem areas that are exposed and become the estuary of waste, especially wastewater pollutants from industrial and urban wastewater that enter the river, and leading to estuary into the sea. The estuary area is a coastal area with mangrove ecosystem. Surabaya city has coastal and marine areas are broad, which rivers and estuary ecosystem which is formed by a mixture of fresh water and sea water with a unique ecosystem and water salinity fluctuates. This ecosystem is called the mangrove ecosystem, which has a flow of water with nutrients important impact on the productivity of coastal areas. However, mangrove areas near to industrial and rural is an area that often polluted from industrial and domestic wastewater.

Acanthus ilicifolius is chosen because this plant live

Rony Irawanto 1*)

1)

Environmental Enginnering Student, Institute Technology of Sepuluh Nopember (ITS), Surabaya, Indonesia

*)

Corresponding author: rony001@lipi.go.id

Sea Holy (Acanthus ilicifolius)

Viability Test on Heavy Metal: Lead (Pb) and

Cadmium (Cd)

The 5 International Conference on Global Resource Conservation

February 12-13

2014

98

|

Batu, East Java, Indonesia

ini mangrove ecosystem, that make environmental balance, but not many reasearch studies looking at the viability this plant to clear wastewater pollutants. The selection of this species based on locallity or location which is a native species, naturally found in nature and some certain conditions releted in phytotechnology. Therefore, testing the viaability of Acanthus ilicifolius to heavy metal contaminants interesting to do. This study aims to determine the concentration of waste pollutant from heavy metals: lead (Pb) and cadmium (Cd), which cause to the death of Acanthus ilicifolius. The information generated is expected to be the basis of phytoforensic research and develop knowledge about the potential plants diversity in the environment phytotechnology.

II. METHOD

This reasearch is a preliminary study to trace concentrations of waste pollutants on plants in phytotechnology. The study was conducted on 20 to 27 December 2013 and 7 to 14 January 2014, with laboratory scale at greenhouse and laboratory Environmental Sanitation and Phytotechnology (SILFI/Sanitasi Lingkungan dan Fitoteknologi) Environmental Engineering Department - ITS. Acanthus ilicifolius are native plants, natural found in estuary and aquatic plants type emergent that have roots, stems and leaves are clear part. Plants were placed in a cylindrical plastic container with a diameter of 8 cm and a height of 15 cm as shown in Figure 1.

Figure 1. Acanthus ilicifolius experimental models in the Range Finding Test (R1= replication first, R2= replication second, R3= replication third).

Planting medium used sand and water, with a quantity of 500 grams sand and 100 ml water, with a 100 ml fertilizer two weeks before exposure. Waste pollutants made artificial such as heavy metals derived from a standard solution of Pb and Cd. Testing with Range Finding Test (RFT) to establish pollutant concentration that causes death in plants. RFT conducted with three replications of two phases, the first phase at a concentration of 500, 1000, 1500, and 2000 mg/L for Pb and Cd, while the second phase at a concentration of 4000, 6000, 8000, and 10,000 mg/L for lead (Pb) and concentration 100, 200, 300, and 400 mg/L for cadmium (Cd). Observations RFT for 7 days (a week) exposure, if not show the changes continued for 7 days (a week) forward. Data obtained and then discussions with outlines and process based on the literature. This are presented in Table and Figure.

III. RESULT AND DISCUSSION

Heavy Metals

Metals are classified as heavy metal when weighing up to 5 grams or more on 1 cm3, so that it can be said that the heavy metal weighs 5 times greater than the weight of water [3]. According to the characteristics of heavy metals classified as having the specific gravity greater than 4 (Palar, 1994) or 5 [6].

Heavy metals can be divided into two categories. Toxic heavy metal, that heavy metal which is not at all needed in the body of living organism, because in small concentrations already toxic, such as Pb, Cd and Hg; and essential heavy metal, which is a heavy metal that is needed by the body of living organism in limited concentration and generally small, becouse that heavy metal use to growth and development of body cells as well as in biochemical processes. If the needs of a very small amount is not met then it would be fatal for the survival of living organism, but if the amount over can be toxic to the body, such as Cu, Zn, Ni, and Cr.

Lead in scientific language called Plumbum and symbolized by Pb. Pb in the group IVA metal group of chemical elements in the periodic table, has an atomic number 82 with an atomic weight of 207.2 [7]. Metallic bluish gray, with high density, soft metal properties, the active chemical properties that can be used to coat metals and low melting point (327.5°C).

Pb is multi-media pollutant for environmental. Among the toxic heavy metals, Pb have distribution wide and primary cause of environmental pollution as well as health problems [8, 9]. Pb is a heavy metal that physiological not need by plants and animals, and can not be degraded. So remediation of heavy metal contamination is done with physically, chemically and biologically [10].

Pb can enter the plant tissue through the leaf surface and root system. While in animals and humans, Pb entry through food and drink as well as through breathing and skin penetration. Pb accumulation in the human body will cause a variety of health effects, including bone loss, reproductive system damage, brain damage, central nervous system toxicity [11], gives the effect on enzyme activity [1], the synthesis hemoglobin, the nervous system, urinary system, reproductive system, and endocrine system [7].

Cadmium (Cd) is a soft metal, silvery white, has the atomic number 48 with an atomic weight of 112.41, is heat resistant and highly resistant to corrosion. So it is good to mix the manufacture of ceramics, enamel, plastic and coating the iron and steel plate [3]. The use of Cd and chemistry compounds commonly found in the dyeing industry, photographic, photoelectric, photoconductor, and others [12]. In addition it is also widely used in small industry such as processing of bread, fish, food, beverages, textiles and others [7].

Cd including heavy metals are dangerous because high risk for blood vessel. Cd effect on humans in the longterm and can accumulate in the body, especially the liver and kidneys. Similarly with Pb, Cd can cause poisoning chronic and acute, which gives effect to the bones, lungs, blood, heart, and reproductive system [7].

Toxic heavy metals such as Pb and Cd can enter the human body through the digestive tract / gastrointestinal

February 12-13

2014

The 5 International Conference on Global Resource Conservation / ingestion, lung/inhalasi, skin / other topical and

parenteral path [13]. Plants have the ability to absorb heavy metals from the environment [14]. Cd more easier accumulated in plants than Pb. Experiments on the absorption of natural vegetation to Pb and Cd concentration was 8 mg/L [15]. On leaf toxic concentrations of Cd and Pb is 5-30 ppm and 30-300 ppm [8].

According to Regulation Law, Surabaya Act No. 2 Year 2004 on water quality management and water pollution control, East Java Governor Act No. 413 Year 1987 on management and water quality in East Java, Goverment No. 82 Year 2001 on management of water quality and water pollution control, Pb and Cd concentration limits that are accept in the environment is 0.03 mg/L and 0.01 mg/L. While Environmetal Ministry Act No. 51/ Kep-MENLH/1/1995 on standar industrial wastewater, Pb concentrations required 0.1 mg/L.

In fact, in Kenjeran channel Surabaya containing Cd average of 5.89 ppm in sediment and 0.09 ppm on water [16]. In research [17] this channel Cd: 0.02 mg/L in water and 3.80 mg/L in the sediment, while in the estuary 0.03 mg/L on seawater and 0.48 mg/L in sediment as well as 0,21 mg/L in the organism (shellfish).

Plant Acanthus ilicifolius (Seaholy)

This plant call Deruju in local name (Indonesia), with synonyms is Acanthus volubilis Wallich (1831). Acanthus plant are herb, erect or reclining shrubby, with mostly perennial herbs. Leaves decussate, simple, undulate to pinnatifid, rarely entire, margins often spiny, rigid, without cystoliths, dark green and shiny; petiole present; stipules absent, but often with a pair of spines from the leaves in the stipular position. Inflorescence a terminal uninterrupted spike. Flowers bisexual, asymmetrical; bracts imbricate, ovate, large; bracteoles in 2 pairs, oblong; calyx 4-partite, lobes imbricate, 2 outer ones larger, corolla tube short, horny, upper lip absent, lower lip elongate, obtusely 3-lobed; stamens 4, slightly didynamous, filaments stout, attached to the corolla throat, anthers 1-celled, linear oblong, bearded along 1 margin; ovules 2 in each loculus, style slender, stigma 2-fid. Fruit a capsule, erect, oblong to ellipsoid. Seeds 4, orbicular, muricate, glandular. Seedling with hypogeal germination. Morphological habitus can be seen in Figure 2.

Figure 2. Acanthus ilicifolius habitus and illustrations.

Grown in Asia tropic and Africa to northern Australia, often found from southern India and Sri Lanka to Indo-China and Indonesia, especially in Java and Madura. Habitat along the estuary of the river and lake, waterfront, commonly found in soil and mangrove

swamps close to the beach. Able to grow to a height 500 m above sea level[18, 19, 20].

Acanthus ilicifolius viability with Pb and Cd

Acanthus ilicifolius know the viability from Range Finding Test (RFT) with done in two stages. RFT first determined the same exposure concentration for Pb and Cd are: 500, 1000, 1500, and 2000 ppm. in 6 days all plant dead for Cd concentration at all, so for the second RFT use concentrations of Cd below 500 ppm. In contrast to Pb concentrations all plant are life, so continued observation for two weeks, and did not any change and death. Therefore the second RFT for Pb concentrations above 2000 ppm.

In the second RFT within 6 days plants dead for Pb concentrations at 8000 and 10,000 ppm, whereas for concentrations Cd at 300 and 400 ppm. Plants that live on Pb and Cd continued observation for two weeks if did not show signs of death. So that the concentration of Pb and Cd deadly to Acanthus ilicifolius at 8000 ppm and 300 ppm, more details can be seen in Table 1.

TABLE 1.

THE CONCENTRATION OF PB AND CD EXPOSURE

Concentration Pb Cd

0 ppm (control) 0 0

RFT 1

500 ppm 0 100

1000 ppm 0 100

1500 ppm 0 100

2000 ppm 0 100

RFT 2

100 ppm - 0

200 ppm - 0

300 ppm - 100

400 ppm - 100

4000 ppm 0 -

6000 ppm 0 -

8000 ppm 100 -

10000 ppm 100 -

Note: 1 ppm (part per miliar) = 1 mg/L.

- = not test, 0 = life, 100 = death.

Acanthus ilicifolius plant used or test have the same criteria. Derived from vegetative propagation (cuttings) trunk diameter 0.5-0.8 cm, age 40-50 days, 15-20 cm height, root length> 10 cm, and the number of leaves 2-6. Observation of Pb and Cd exposure waste as can be seen in Table 1 Acanthus ilicifolius change morphologically. Showed morphological changes to the death of the leaves are not erect, the plant looks weak stem, and leaf discoloration. The changes can be seen in Figure 3.

The 5 International Conference on Global Resource Conservation

February 12-13

2014

100

|

Batu, East Java, Indonesia

The morphological changes also supported with observation of high plants and number of leaves that live after a week exposure. Observations on the second RFT, can see in Table 2.

TABLE 2.

THE NUMBER OF LEAVES AND HEIGHT PLANT Concentration Test 0 day 6 day

Pb

4000 ppm R1 10 cm 16 cm 6/6 leaf 6/6 leaf R2 12 cm 17 cm 4/6 leaf 6/8 leaf R3 15 cm 21 cm 6/4 leaf 8/6 leaf

6000 ppm R1 10 cm 16 cm 6/6 leaf 6/6 leaf R2 13 cm 18 cm 6/6 leaf 8/6 leaf R3 16 cm 21 cm 6/0 leaf 10/0 leaf

Cd

100 ppm R1 14 cm 14 cm 0/6 leaf 0/6 leaf R2 12 cm 15 cm 3/4 leaf 6/6 leaf R3 18 cm 18 cm 0/6 leaf 0/6 leaf 200 ppm R1 12 cm 12 cm

6/4 leaf 6/4 leaf R2 14 cm 13 cm 4/4 leaf 4/4 leaf R3 21 cm 20 cm 6/0 leaf 6/0 leaf Note: R = replication - , a/b leaves

a = branch right, b = branch left.

Figure 4. Stem and leaves of Acanthus ilicifolius are discolorated compared with control .

Based on Table 2 known that concentrations exposure that are not lethal, for Pb are increase the growth. It seen from the height of branch / stems and number of leaves. Whereas for Cd concentrations, can inhibit the growth of branch stems and leaves. Besides morphological changes in Acanthus ilicifolius plants heavy metal exposed, lead (Pb) and cadmium (Cd) can be easily seen, as in Figure 4. Pb exposure on leaf veins and stems brownish-black to blackish green leaves. While Cd exposure in the leaves more pale green to yellowish.

IV. CONCLUSION

Result show that Acanthus ilicifolius (seaholy) death in letal concentration for 8000 mg/L in lead (Pb) and 300 mg/L in cadmium (Cd). Cd concentrations although not lethal but inhibit the growth of stems and leaves, in other hand Pb concentrations did not show inhibit indication but accelerating growth. In addition to the color change of Pb exposed to brownish-black on leaf veins and blackish green on stem, but Cd exposed that green pale to yellowish before die.

Suggestion: it is necessary for further research on heavy metal concentrations of Pb and Cd contained in each part of plants such as leaf, steam and root.

ACKNOWLEDGMENT

The author would like to thank the ministry of research and technology for scholarship and support to participation in any scientific activities, like this. The authors also thank to Dr. R. Hendrian, MSc. and Prof. Dr. Ir. Sarwoko Mangkoedihardjo, MSc.ES for suggestion and discussion that very useful.

REFERENCES

[1] Widowati, W., Sastiono, A., and Rumampuk, R.J. 2008. Effects Metals Toxic: Prevention and Control Pollution. ANDI Publisher. Yokyakarta.

[2] Wang, L.K., Chen, J.P., Hung, Y.T. and Shammas, N.K. 2009. Heavy Metals in the Environment. CRC press. New York.

[3] Darmono. 1995. Metals in Biological Systems of Organism. Indonesia University. Jakarta.

[4] Mangkoedihardjo, S., and Samudro, G. 2010. The Applied Phytotechnology. Graha Ilmu. Yogyakarta.

[5] Mangkoedihardjo, S., and Samudro, G. 2009. Ecotoxicology Technosphere. Guna Widya. Surabaya.

[6] Soemirat, J. 2003. Environmental Toxicology. Gadjah Mada University Press. Yokyakarta.

[7] Palar, H. 1994. Contamination and Toxicology of Heavy Metal. Rineka Cipta. Jakarta.

[8] Allowey, B.J. and Ayres, D.C. 1997. Chemical Principles of Environmental Pollution. 2nd Edition. Blackie Academic and Professional Chapman and Hill. London.

[9] Cunningham, S.D. and Berti, W.R. 1993. Remediation of Contaminated Solid With Green Plants: an overview. In Vitro Cell. Dev. Biol. 29 . page 207-212.

[10] Hindersah, H., Kalag, M.A. and Muntalif, S.B. 2004. Accumulation of Pb and Cd Tomato Fruit that Planted on Dry Land Mud Containing of Domestic Wastewater Treatment. Environmental Engineering. ITB. Bandung.

[11] Amriani. 2011. Bioaccumulator Heavy Metal: Lead (Pb) and Zinc (Zn) On The shell blood (Anadara Granosa L.) and The shell mangrove (Polymedosa bengalensis L.) in the waters of Kendari Bay. Environmental Science Thesis. Diponegoro University. Semarang.

[12] Darmono, 2006. Environment and Pollution: relationship with toxicology of metal compounds. Indonesia University. Jakarta.

Control steam

February 12-13

2014

The 5 International Conference on Global Resource Conservation

Batu, East Java, Indonesia

|

101

[13] Ahmad, R. 2004. Environmental Chemistry. ANDI

Publisher. Yogyakarta.

[14] Irwan A., Noer, K., and Yenny E. 2008. Studies absorption Metal Cd, Ni, and Pb Concentration With Variations In Roots, Stems and Leaves of Spinach Plants. Lambung Mangkurat University, Banjarmasin.

[15] Neis, U. and Bittner, A. 1989. Wastewater Use. Obor Indonesian Foundation. Jakarta.

[16] Lembahingtyasadjeng, F.H. 2003. Review of Heavy Metal Cadmium (Cd) at Kenjeran Channel Effect of Industrial Waste and Strategy Management. Environmental Engineering Thesis. ITS. Surabaya.

[17] Hasan, I. 2005. Dynamic Modeling System for Regional Ecosystem Management of Kenjeran Coastal for Heavy Metal Pollution of Cadmium (Cd). Environmental Engineering Thesis. ITS. Surabaya.

[18] Hidayat, Yuzammi, S., Hartini, S., and Astuti, I.P. 2004. Aquatic Plants of Bogor Botanic Garden (Vol.1 No.5). Bogor Botanic Garden. Bogor.

[19] Backer, C.A. and Bakhaizen, R.C. 1963. Flora of Java. The Rijksherbarium. Netherlands.

[20] vanValkenberg, J.L.C.H., and Bunyapraphatsara. 2002. Plant Resources of South-East Asia No. 20 (2): Medical and Poisoning Plant 2. PROSEA Foundation. Bogor.

The 5 International Conference on Global Resource Conservation

February 12-13

2014